多重模糊条件下的水上服务区类型决策

Research for Type Selection of Waterway Service Area under Multiple-fuzziness

针对水上服务区类型选择的多重模糊性问题,提出基于模糊贝叶斯的决策方法,根据文献确定水上服务区选择的影响因素和决策准则,建立四层决策框架;根据各服务区的供需关系对影响因素进行模糊化,利用改进的IF-THEN规则建立影响因素和输入变量之间、影响因素和决策准则之间的推理关系;利用贝叶斯网络对推理规则库进行转换,通过计算贝叶斯网络的边缘概率获得决策方案值;对各节点的决策状态赋予效用值及权重,获得各水上服务区类型的综合得分。将该方法应用于芜申运河宜兴水上服务区的类型选择,结果表明,该方法能够很好地实现水上服务区的类型选择。

Aiming at the multiple-fuzziness problems in selecting the appropriate type of inland waterway service area,a decision-making method is proposed based on fuzzy Bayesian network. From the literature review,the influencing factors of inland waterway service area were obtained and the decision attributes were also defined to establish the four-layer decision-making framework. The membership function was introduced to fuzzify the influencing factors,and the improved IF-THEN rule was used to setup the relationship between influencing factors and input variables and influencing factors and attributes respectively. After transformation the rule base into Bayesian network,the decision matrix was obtained by the marginal probability. The utility value was introduced to obtain the utilities of each functional type. The inland waterway service area of Yixing is used to verify the proposed approach as an illustrative example,showing that this model is useful and beneficial for type selection of waterway service area.